An improved clustering algorithm and its application
| Autor: | Yunhong Ma, Xue Yang, Feng Chen, Chenghan Wang |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Fuzzy clustering
General Computer Science business.industry Computer science Single-linkage clustering Correlation clustering Pattern recognition computer.software_genre Determining the number of clusters in a data set ComputingMethodologies_PATTERNRECOGNITION Data stream clustering CURE data clustering algorithm Canopy clustering algorithm Artificial intelligence Data mining Electrical and Electronic Engineering business Cluster analysis computer |
| Zdroj: | International Journal of Wireless and Mobile Computing. 12:358 |
| ISSN: | 1741-1092 1741-1084 |
| Popis: | Cluster analysis is an important issue for machine learning and pattern recognition, and clustering algorithms are used to solve these problems. Owing to the disadvantage of the traditional clustering algorithms that need iteration calculation and do not adapt to clustering random distribution data, an improved clustering algorithm named automatic clustering algorithm based on data contained ratio is developed. The concept of data contained ratio is proposed in the improved algorithm, the number of clusters is automatically determined based on data contained ratio, and the cluster centres are obtained accordingly. Several group test data testify and demonstrate the validity and effectiveness of the improved clustering algorithm. In addition, the comparison between the traditional K-means clustering algorithm and improved clustering algorithm is presented. Finally, the improved clustering algorithm is used to recognise the flight trace. The results demonstrate that the improved clustering algorithm can be used for pattern recognition; it has high adaptability and robustness in clustering random distribution data set. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|